You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello! I am using rl-zoo to train a manipulator to perform contact tasks. I want to compare the agents based on the total contact forces produced during the evaluation episodes. How would you log this metric?
The force values are computed by the environment and included in the info dictionnary at each step. My first idea was to modify ExperimentManager to use MyCustomEvalCallback instead of EvalCallback but then I believe I would also need to implement a custom evaluation_policy to replace the one from SB3 (stable_baselines3.common.evaluation.evaluate_policy).
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello! I am using
rl-zoo
to train a manipulator to perform contact tasks. I want to compare the agents based on the total contact forces produced during the evaluation episodes. How would you log this metric?The force values are computed by the environment and included in the
info
dictionnary at each step. My first idea was to modifyExperimentManager
to useMyCustomEvalCallback
instead ofEvalCallback
but then I believe I would also need to implement a customevaluation_policy
to replace the one from SB3 (stable_baselines3.common.evaluation.evaluate_policy
).Any other ideas on how to implement this?
PS: Thanks for the awesome library
Beta Was this translation helpful? Give feedback.
All reactions